Surgical resection of glioblastomas (GBM) fails to completely remove invasive tumor tissue, resulting in tumor recurrence. Virus vector delivery of anti-tumor genes to remaining malignant cells has potential to prevent this recurrence. An enduring issue with virus vectors is inefficient transgene delivery throughout the tumor after direct injection. As the brain and brain tumors are both highly vascularized, delivery of therapies via the vasculature is an attractive alternative approach for widespread transgene delivery. However, current vector systems are suboptimal for intravascular (i.v.) gene delivery to the brain due to liver uptake, the impenetrability of the blood-brain barrier (BBB), and pre-exisiting immunity to the virus (e.g. neutralizing antibodies to the virus capsid). We therefore sought alternative strategies which would serve two purposes: (1) to allow specific targeting of adeno-associated virus (AAV) vector to the brain tumor environment (BTE) and (2) shield AAV capsids from neutralizing antibodies. Microvesicles (MV) are naturally secreted, membrane-encompassed structures known to deliver proteins, RNA, miRNA, and DNA to neighboring cells. Accumulating research suggests they may be suitable for targeted gene/protein transfer. We have discovered that during vector production, a portion of AAV isolated from the media is associated with MV. AAV capsids were localized primarily on the inside of MV by transmission electron microscopic examination. We have found that the MV-associated AAV, termed MV-AAV, are more efficient at gene delivery to cultured cells compared to standard cell lysate purified AAV vectors. We have also shown that surface tagging MV-AAV with magnetic nanoparticles allows for directional AAV-mediated gene transfer via the use of a magnet placed on the bottom of the cell culture well. Furthermore a recent study showed that MV's loaded with siRNA could be specifically targeted to the brain after i.v. injection in mice. This proposal will test the hypothesis that MV-AAV can be targeted to the brain tumor environment after i.v. injection to therapeutically treat the tumor. This work has potential to provide a non-invasive means of robust and global gene delivery to the brain.